Consumption of isoflavone-rich soy has been suggested to lower prostate cancer risk. The soy components thought to be responsible for this effect are the isoflavones. As a result, soy isoflavones are currently being investigated in epidemiological and clinical studies for their possible roles in prostate cancer prevention. It is extremely important in these studies to have an accurate estimate of isoflavone exposure in individuals. Given the inadequacy of dietary methodologies, biological fluids are considered more reliable indicators of isoflavone exposure. In designing these studies, researchers have generally assumed that plasma and 24- hr urine isoflavone concentrations accurately reflect prostate tissue exposure. However at this time, no information is available on the validity of this assumption. Thus, the main objective of this application is to evaluate the reliability of plasma and urinary isoflavone levels by comparing blood and urine concentrations to more direct measurements of prostate exposure. To accomplish this, the proposed study will compare isoflavone levels in plasma and 24-hr urine with those in expressed prostatic secretion (EPS), a widely accepted marker of prostatic tissue exposure. In addition, isoflavone levels in semen and post-massage urine will be determined and compared with EPS, in order to determine whether these fluids are more accurate than blood or urine and more easily obtainable than EPS in clinical or epidemiological studies. The specific aims of this project are to assess concentrations of isoflavones (genistein, daidzein, equol, ODMA, glycitein and dihydrodaidzein) in blood, 24-hr urine, semen, post-massage urine and EPS after soy isoflavone consumption. A randomized double-blind parallel arm study will be performed in 50 young men assigned to receive a tablet containing either 30 or 60 mg of isoflavones for a period of four days. Before and after the treatment period phytoestrogen concentrations 24-hr urine, plasma, EPS, post-massage urine and semen will be measured. Comparing 24-hr urine, plasma, post-massage urine and semen concentrations with those in EPS will enable us to assess the validity of blood, urine, and semen as predictors of prostate tissue phytoestrogen exposure. Comparing the two treatment groups will allow us to assess dose-dependant effects on phytoestrogen distribution. Although EPS is easier to obtain than prostate tissue,,the procedure is uncomfortable and requires trained personnel. If semen orposf-massage urine concentrations correlate well with EPS, these collections would be more practical for estimating tissue exposure in epidemiological and intervention studies. If blood and 24-hr urine concentrations correlate well with EPS, their validity will be confirmed and we will show that collection of the other fluids is unnecessary. The results of this study will help determine which biological fluid best reflects prostate exposure to isoflavones and may help improve the design of future epidemiological and intervention studies on isoflavones and prostate cancer risk reduction. [unreadable] [unreadable] [unreadable]